diff --git a/docs/hygiene-history/ticks/2026/05/01/0953Z.md b/docs/hygiene-history/ticks/2026/05/01/0953Z.md
new file mode 100644
index 000000000..8f482bee6
--- /dev/null
+++ b/docs/hygiene-history/ticks/2026/05/01/0953Z.md
@@ -0,0 +1 @@
+| 2026-05-01T09:53:00Z | opus-4-7 / autonomous-loop tick | 98fc7424 | Stale-state + deferred-per-policy tick — checked next-batch of older PRs. (1) **PR #1025 (orthogonality-rule memory) — stale-count, all threads already resolved**. The list-summary view showed 1 unresolved but per-thread query returned `isResolved: true` on all 5 threads (all Codex/Codepilot stale-filename findings on the orthogonality-rule memory file's composes-with chain — all already addressed in prior work). PR #1025 is in CI-settling state pending host's auto-merge automation; no action needed. (2) **PR #755 (2026-04-29 prefab tick-history shard, 1 thread)** — Codex P1 finding about timestamp mismatch (shard claims `2026-04-29T05:00:00Z` but commit was at `2026-04-29T02:39:43Z`). This is the **prediction-vs-receipt-column class** Aaron 2026-05-01 explicitly classified as low-stakes greenfield-OK in B-0129's filing: *"at this point if history is not perfect it's okay clean it up if you like, if not just note it somewhere and lets do it right in the future, still low steakes, up to you greenfield."* B-0129 acceptance criteria item #2 states the migration policy: *"leave the 14 existing prefab shards from 2026-04-29 [as-is]"*. Resolved thread with explicit deferred-per-B-0129-migration-policy reply. This is a **new resolution-class: deferred-per-explicit-Aaron-policy**. Cron 98fc7424 healthy. | [PR #755: 1 thread resolved with B-0129-policy deferral reply (no edit); PR #1025: stale-count audit confirmed all-resolved] | The deferred-per-explicit-Aaron-policy class is the right shape for findings on **frozen historical artifacts** — when reviewer's finding is technically correct but the underlying substrate has been explicitly classified as low-stakes-don't-cleanup by the maintainer, the right resolution is reply-with-policy-citation + resolve, NOT edit. The substrate-author-intent (prefab shards exist as historical artifacts of the 2026-04-29 prefab-cluster experiment) is preserved precisely BY not retroactively rewriting it. Future-Otto: when a finding lands on substrate explicitly classified as historical/low-stakes/greenfield-OK by an existing memory or backlog row, cite that classification when resolving. The class is structurally similar to "outdated thread" but distinct: outdated means subsequent commits made the finding moot; deferred-per-policy means the finding STANDS but Aaron has explicitly traded the cleanup against other priorities. The stale-count audit on PR #1025 is also a class-level lesson: **list-summary thread counts can lag per-thread state** — for trustworthy thread-status, query per-PR via GraphQL with explicit `isResolved` filter rather than trusting the summary count. |